Prognosis in cutaneous T-cell lymphoma by skin
`stage: Long-term survival in 489 patients
`
`Herschel S. Zackheim, MD,a Smita Amin, MD,a* Mohammed Kashani-Sabet, MD,a and
`Alex McMillan, PhDb San Francisco, California
`
`Background: Although a number of studies have documented the long-term survival of
`patients with cutaneous T-cell lymphoma (CTCL), none have provided data as to the rela-
`tive survival of all 4 skin stages.
`Objective: We document survival of CTCL patients by T stage relative to that of an age-,
`sex-, and race-matched population.
`Methods: The survival of 489 patients with CTCL registered since 1957 was compared
`with that of a California control population.
`Results: For stage T1 (< 10% skin involved) there was no significant difference between
`the observed and expected survivals. For the other 3 stages the observed survival was sig-
`nificantly inferior to that of the expected survival (P = .002). At 10 years the relative sur-
`vivals were: T2 (10% or more skin involved) 67.4%, T3 (tumor stage) 39.2%, T4 (gener-
`alized erythroderma) 41.0%. T2 plaque stage patients had an inferior relative survival (P =
`.001), whereas T2 patch stage patients did not. Lymphadenopathy had an unfavorable
`impact on prognosis. There was a strong trend toward diagnosing CTCL at an earlier stage
`in more recent years. We estimate that from 15% to 20% of our patients died of CTCL or
`related complications.
`Conclusion: The relative survival of CTCL patients worsens with increasing skin stage,
`although stages T3 and T4 had closely similar survivals. The great majority of patients
`with CTCL do not die of their disease.
`(J Am Acad Dermatol 1999;40:418-25.)
`
`Primary “classic” cutaneous T-cell lymphoma
`(CTCL) (mycosis fungoides (MF)/Sézary syn-
`drome) is an epidermotropic lymphoma whose ini-
`tial clinical manifestations are in the skin.1 Other
`primary CTCLs that are predominantly nonepider-
`motropic, such as CD30+ or CD30– large T-cell
`lymphoma, pleomorphic CTCL, subcutaneous T-
`cell lymphoma, and CTCLs caused by systemic T-
`cell lymphomas are not included in this study.
`Henceforth, in this report CTCL refers only to
`classic CTCL.
`
`the Cutaneous Oncology Division, Department of
`From
`Dermatology,a and the Biostatistics Core, Cancer Center,b
`University of California, San Francisco.
`Dr Kashani-Sabet is supported by the Leaders Society Clinical Career
`Development Award of the Dermatology Foundation, and the
`Herschel and Diana Zackheim Endowed Fund.
`Accepted for publication Nov 6, 1998.
`Reprint requests: Herschel S. Zackheim, MD, 133 Arch St, Redwood
`City, CA 94062.
`*Present address: Division of Dermatology, The Toronto Hospital-
`Western Division, University of Toronto, Ontario, Canada.
`Copyright © 1999 by the American Academy of Dermatology, Inc.
`0190-9622/99/$8.00 + 0 16/1/95665
`
`418
`
`The relationship between disease stage and sur-
`vival in cancer patients is well recognized. A num-
`ber of studies have analyzed the relationship
`between CTCL stage and survival. Four of these
`involved at least 100 patients followed up for 5
`years or more and were analyzed according to
`extent and character of skin involvement as well as
`other parameters.2-5 However, with the exception of
`the study of Kim et al,4 which compared the
`observed versus the expected survival of patients
`with limited extent disease, none of the reports have
`compared survival of CTCL patients with that of a
`control population. In this study we report the sur-
`vival of 489 patients with CTCL stratified according
`to skin stage and compared, according to stage, with
`an age-, sex-, and race-matched control population.
`Additional analyses were made of the possible
`influence of depth and type of skin infiltrate (patch
`vs plaque), lymph node, and blood status.
`
`METHODS
`Patients were registered at the cutaneous lymphoma
`clinic of the University of California, San Francisco, the
`
`IPR2018-00685
`Celgene Ex. 2011, Page 1
`
`

`

`Journal of the American Academy of Dermatology
`Volume 40, Number 3
`
`Zackheim et al 419
`
`Table I. TNM classification of cutaneous T-cell
`lymphoma (CTCL)*
`
`Table II. Demographics of 489 patients with
`CTCL
`
`T: Skin
`T1 Patches, papules, or plaques involving < 10% of
`skin
`T2 Patches, papules, or plaques involving 10% or
`more of skin
`T3 Tumors (one or more)
`T4 Generalized erythroderma
`N: Lymph nodes
`N0 No clinically abnormal peripheral lymph nodes,
`pathology negative for CTCL (if biopsy obtained)
`N1 Clinically abnormal peripheral lymph nodes,
`pathology negative for CTCL (if biopsy obtained)
`N2 No clinically abnormal peripheral lymph nodes,
`pathology positive for CTCL
`N3 Clinically abnormal peripheral lymph nodes,
`pathology positive for CTCL
`B: Blood
`B0 No evidence of increased numbers of atypical cir-
`culating cells
`B1 Evidence of increased numbers of atypical circu-
`lating cells
`M: Visceral organs
`M0 No evidence of visceral involvement
`M1 Visceral involvement confirmed by pathology
`
`*Modified from Bunn and Lamberg.7
`
`Veterans Administration Medical Center, San Francisco,
`or at the private office of one of us (H. Z.).
`Histologic diagnoses were made in accord with pub-
`lished criteria.6 Categorization as to patch or plaque
`stage disease was also made on the basis of histologic
`features.6 Patients were classified according to the
`tumor-node-metastasis (TNM) system7 (Table I). All
`patients received a minimum work-up of a complete
`skin examination, palpation for lymph nodes, complete
`blood cell count, chest x-ray, and blood chemistries for
`liver and kidney function. Estimation of the extent of
`skin involvement was usually made by at least 2
`observers using the “rule of 9s”8 and the fact that the
`palmar surface represents about 1% of the skin surface.7
`Computed tomographic scans were obtained in selected
`patients with palpable lymphadenopathy, erythroderma,
`or tumors. Lymph node biopsy specimens were
`obtained in those with a lymph node at least 2 cm in
`diameter or with multiple nodes at least 1 cm in diame-
`ter. Examination of the blood for evidence of Sézary
`cells was obtained in all patients with generalized ery-
`throderma, and in selected patients with widespread
`plaque or tumor stage disease or significant adenopathy.
`Bone marrow biopsy specimens were obtained in
`selected patients with advanced disease.
`In accord with Lamberg et al,2 survival was calculat-
`ed from the date of registration to death or last follow-
`
`No. (%)
`
`T1
`
`T2
`
`T3
`
`T4
`
`11 (2.2)
`25 (5.1)
`54 (11.0)
`62 (12.7)
`101 (20.7)
`116 (23.7)
`96 (19.6)
`24 (4.9)
`489 (100)
`190 (38.9)
`299 (61.1)
`
`20 (4.1)
`48 (9.8)
`421 (86.1)
`
`4
`11
`28
`28
`39
`39
`21
`4
`174
`70
`104
`
`5
`6
`163
`
`5
`11
`18
`27
`39
`44
`47
`8
`199
`81
`118
`
`11
`33
`155
`
`1
`2
`6
`3
`7
`14
`11
`3
`47
`15
`32
`
`0
`6
`41
`
`1
`1
`2
`4
`16
`19
`17
`9
`69
`24
`45
`
`4
`3
`62
`
`Age (y)
`< 20
`20-29
`30-39
`40-49
`50-59
`60-69
`70-79
`80+
`Total
`Female
`Male
`Race
`AP
`Black
`White
`
`AP, Asian Pacific.
`
`up. Calculation from time of registration, rather than at
`time of histologic diagnosis, is necessary to ensure uni-
`formity in criteria for staging. Approximately 85% of
`the patients were registered within 3 months of histo-
`logic diagnosis. Patients still living at the time of last
`follow-up were considered as censored observations.
`Deaths from all causes were used in the analysis.
`Survival probabilities were calculated according to the
`method of Kaplan and Meier. Differences in survival
`between different subgroups were evaluated by means
`of the log-rank test. Multivariate adjustment for poten-
`tial confounders (eg, age) were made by means of the
`Cox proportional hazard model.
`Expected survival of patients was calculated using
`the 1990 published life tables for California.9 (Earlier
`life tables were not available.) For each year n from
`time of registration, the probability of surviving 1 addi-
`tional year was calculated for each patient still at risk,
`and the expected survival at year n was obtained by
`multiplying the expected survival at year n-1 by the
`average survival probability of those at risk.
`
`RESULTS
`The cohort included 489 patients with CTCL
`registered since 1957 (Table II). The cut-off date
`for entry in the study was Dec 31, 1994. The cut-
`off date for follow-up observation was Dec 31,
`1995. Only 10 patients were registered before
`1971, the year that one of us (H. Z.) began to see
`patients with CTCL. The age range at registration
`was 510 to 93 years (mean, 56.5; median, 59.0
`
`IPR2018-00685
`Celgene Ex. 2011, Page 2
`
`

`

`420 Zackheim et al
`
`Journal of the American Academy of Dermatology
`March 1999
`
`Table III. Observed and expected survival according to T stage
`
`Stage
`
`Year
`
`Sig
`
`Observed survival
`(%) CI
`
`Expected survival
`(%)
`
`Relative survival*
`(%)
`
`T1
`
`T2
`
`T3
`
`T4
`
`5
`10
`15
`5
`10
`15
`5
`10
`15
`5
`10
`15
`
`NS
`NS
`NS
`P < .001
`P = .002
`P < .001
`P < .001
`P < .001
`P < .001
`P < .001
`P < .001
`n/a
`
`94.8 (90.9-98.6)
`83.1 (75.1-91.1)
`79.5 (70.3-88.6)
`75.7 (68.8-82.6)
`55.2 (46.0-64.3)
`47.1 (36.9-57.3)
`45.0 (30.0-60.0)
`28.9 (14.1-43.8)
`21.7 (7.6-35.8)
`50.6 (38.0-63.3)
`29.7 (16.2-43.2)
`n/a
`
`92.2
`83.1
`74.0
`90.4
`81.8
`75.4
`87.4
`73.8
`53.6
`88.4
`72.4
`n/a
`
`CI, 95% confidence interval; n/a, not applicable (there were no 15-year survivors); NS, not significant; Sig, significance.
`*Relative survival is calculated by dividing observed by expected survival.
`
`102.7
`100.1
`107.4
`83.8
`67.4
`62.5
`51.5
`39.8
`40.5
`57.3
`41.0
`n/a
`
`years). Of the registrants, 2.2% were under age 20,
`7.4% were under age 30, and 68.9% were age 50
`or older. There were 299 (61.1%) men and 190
`(38.9%) women (ratio 1.57). White patients
`(including Hispanics) numbered 421 (86.1%),
`blacks 48 (9.8%), and Asian-Pacific 20 (4.1%). Of
`the Asian-Pacific population, 11 were Chinese, 5
`Filipino, 3 East Indian, and 1 undetermined.
`A total of 174 patients were classified as stage
`T1, 199 as T2, 47 as T3, and 69 as T4. Patients
`tended to be older in the more advanced than the
`earlier stages. Thus, according to stage, the mean
`ages were: T1 52.3, T2 56.9, T3 59.4, and T4 64.2.
`Noteworthy is the higher proportion (37.5%) of
`stage T4 in patients 80+. There was no significant
`gender difference as to the proportion in the vari-
`ous stages, or as to the date of registration. Blacks
`comprised 9.8% of the total cohort, which is
`slightly larger than the 7.4% of blacks in
`California9 (P = .04; ratio 1.32). Black patients
`had a relatively more advanced stage than did
`whites. Thus, 87.5% of black patients were in
`either stages T2, T3, or T4 as compared with
`61.3% of whites in those stages. The median fol-
`low-up for the entire cohort of 489 patients was 4.7
`years.
`In Table III we present data as to the observed
`and expected survival at 5, 10, and 15 years
`according to stage, adjusted for age, sex, and race,
`and state the corresponding relative survival
`(observed as a proportion of the expected). Plots of
`the observed and expected survival for each stage
`are shown in Fig 1.
`
`The relative survival of T1 patients was compa-
`rable to the general population; in fact, their prog-
`nosis was somewhat better at 5 and 15 years. This
`may reflect a higher socioeconomic status of our
`patients as compared with the general population.
`The relative survival of the other 3 stages was sig-
`nificantly (P < .001) inferior to that of the general
`population. For T2 the relative survival at 10 years
`was 67.4%. The relative survivals at 10 years for
`T3 and T4 were lower and were not greatly differ-
`ent from each other—T3: 39.2% and T4: 41.0%.
`Because we used 1990 life tables the expected
`survivals may be slightly higher than would have
`been obtained using earlier life tables. Two percent
`of the patients were registered before 1970, 19.2%
`before 1980, and 61.1% before 1990.
`Data were analyzed as to the possible influence
`of patch versus plaque stage on survival (Table IV).
`An analysis of the data for stage T1 is not relevant
`because the survival of those patients is comparable
`with that of the general population. The relative sur-
`vival of patch stage T2 patients is little different
`from that of the general population. However, the
`relative survival for plaque stage T2 patients was
`significantly lower than that for an age-, sex-, and
`race-matched population (P = .002).
`Data were also analyzed as to the possible influ-
`ence of nodal status on survival. This included all
`degrees of nodal abnormality (N1 to N3). After
`adjusting for stage and demographics, nodal
`abnormality had a marginally significant effect on
`survival (P = .03). The estimated hazard ratio (rel-
`ative risk of dying) was 1.48 in patients with nodal
`
`IPR2018-00685
`Celgene Ex. 2011, Page 3
`
`

`

`Journal of the American Academy of Dermatology
`Volume 40, Number 3
`
`Zackheim et al 421
`
`Fig 1. Kaplan-Meier estimates of survival. Dashed lines track expected survival for a
`group of individuals with age, sex, and race characteristics similar to those of our study
`population. Vertical lines are 95% confidence intervals for observed survival. Numbers
`above time axis are number of patients at risk. Tics indicate patients who are still alive. A,
`T1. There is no significant difference between observed and expected survival. B, T2. P =
`.002 that observed survival differs from expected. C, T3. P < .001 that observed survival
`differs from expected. D, T4. P < .001 that observed survival differs from expected.
`
`IPR2018-00685
`Celgene Ex. 2011, Page 4
`
`

`

`422 Zackheim et al
`
`Journal of the American Academy of Dermatology
`March 1999
`
`Table IV. Stage T2 observed and expected survival of patch and plaque stage*
`
`Year
`
`Sig
`
`Observed survival
`(%) CI
`
`Expected survival
`(%)
`
`Relative survival
`(%)
`
`PA
`
`PL
`
`5
`10
`15
`5
`10
`15
`
`NS
`NS
`NS
`P = .001
`P < .001
`P < .001
`
`82.9 (73.4-92.5)
`71.9 (57.1-86.7)
`71.9 (57.1-86.7)
`72.7 (62.6-82.7)
`48.9 (36.9-61.0)
`38.1 (25.4-50.7)
`
`90.6
`81.4
`75.2
`89.3
`80.5
`73.5
`
`91.5
`88.3
`95.6
`81.3
`60.8
`51.8
`
`PA, Patch stage; PL, plaque stage.
`*See footnotes for Table III.
`
`Table V. CTCL stage relative to time of registra-
`tion
`
`Stage
`
`All
`T1
`T2
`T3
`T4
`
`1957-1979
`No. of patients (%)
`
`1990-1994
`No. of patients (%)
`
`94 (100)
`22 (23.4)
`45 (47.9)
`16 (17.0)
`11 (11.7)
`
`190 (100)
`79 (41.6)
`72 (37.9)
`13 (6.8)
`26 (13.7)
`
`abnormality relative to patients without nodal
`abnormality (N0). However, when data were ana-
`lyzed separately by stage, there was a trend for
`nodal status to be more important in the earlier
`stages; the hazard ratio was estimated at 3.0 in T1,
`1.9 in T2, 1.3 in T3, and 0.93 in T4. However, this
`trend was not statistically significant (P = .18), and
`our data do not provide sufficient statistical power
`to detect such a trend. We estimate that more than
`1700 subjects would be required to provide 80%
`power to detect a trend of this magnitude.
`The blood was examined for evidence of Sézary
`cells in 62 patients, and was positive in 50 patients
`and negative in 12. These numbers were too small
`to permit a statement as to possible effect of blood
`involvement on survival. Similarly, the number of
`patients with bone marrow or other visceral biop-
`sies was too small to permit analysis of visceral
`involvement as an independent factor.
`A definite trend for diagnosing CTCL at an ear-
`lier stage in more recent years is evident (Table V).
`Particularly striking, when comparing the period
`of 1957 to 1979 with the period of 1990 to 1994,
`is the marked increase in the percentage of patients
`in T1 from 23.4% to 41.6%, and the decrease in T3
`patients from 17.0% to 6.8%.
`
`Equally striking were the differences between
`those periods as to patients with patch versus
`plaque stage. Thus, in the period of 1957 to 1979,
`of 67 patients (combined T1 and T2) 9 (13.4%)
`were patch stage, 47 (70.1%) were plaque stage,
`and 11 (16.4%) were not classified as either. In
`1990 to 1994, of 151 T1 and T2 patients 122
`(80.8%) were patch stage, 24 (15.9%) were
`plaque stage, and 5 (3.3%) were not categorized
`as either.
`We made an analysis for improvement in prog-
`nosis for time trend. The only significant differ-
`ence was in patch stage T2 where later registrants
`seem to be doing better. There was a nonsignifi-
`cant trend for improved prognosis for plaque stage
`in more recent years.
`Although in our estimate of survival all causes
`of death are included, we also estimate the proba-
`ble cause of death based on available information
`(Table VI).
`The highest proportion of deaths for which the
`cause was not known was in stage T1 (73.7%).
`This is understandable because patients who do
`well are more likely to be lost to follow-up as com-
`pared with those with advanced disease. We esti-
`mated the number of deaths caused by CTCL
`according to the method of Weinstock and
`Reynes.11 According to that method the number of
`deaths attributable to CTCL can be calculated by
`subtracting the number of expected deaths from
`the observed deaths. The excess represents the
`deaths attributal to the disease. The expected
`deaths were determined from the life table for
`California.9
`For stage T1 19 deaths were observed, whereas
`for the age-, sex-, and race-matched control popu-
`lation 24 deaths were expected. Thus, conversely,
`there were 5 fewer deaths than expected, but the
`
`IPR2018-00685
`Celgene Ex. 2011, Page 5
`
`

`

`Journal of the American Academy of Dermatology
`Volume 40, Number 3
`
`Zackheim et al 423
`
`Table VI. Causes of death in patients with CTCL
`
`Cause of death
`
`Deaths
`
`Stage
`
`T1
`T2
`T3
`T4
`Total
`
`No.
`
`174
`199
`47
`69
`489
`
`D
`
`19
`67
`35
`42
`163
`
`R
`
`1
`19
`15
`18
`53
`
`NR
`
`4
`30
`6
`12
`52
`
`UN (%)
`
`PY
`
`14 (73.7)
`18 (26.9)
`14 (40.0)
`12 (28.6)
`58 (35.2)
`
`1,242
`1,279
`271
`300
`3,092
`
`Obs
`
`19
`67
`35
`42
`163
`
`Exp
`
`24.0
`27.4
`10.6
`9.2
`71.3
`
`Excess (%)
`
`–5.0 (0)
`39.6 (19.9)
`24.4 (51.9)
`32.8 (47.5)
`91.7 (18.8)
`
`D, Deaths; Exp, expected; No., number of patients; NR, death not related to CTCL or treatment; Obs, observed; PY, patient-years of follow-up;
`R, death related to CTCL or treatment for CTCL; UN, cause of death unknown. Note: For stage T1 observed and expected deaths do not differ sig-
`nificantly (P = .13). All other differences are significant (P < .001).
`
`difference was not statistically significant (P =
`.13). As previously noted with regard to the rela-
`tive survival curve for stage T1, this may reflect a
`better socioeconomic status of our patients as
`compared with the general population. For all
`other stages the number of observed deaths
`exceeded the expected by a significant margin (P
`< .001).
`The data in Table VI indicate that 91.7 (18.8%)
`of the total cohort of 489 patients are estimated to
`have died of CTCL or related complications.
`Only 2 T1 patients progressed to more
`advanced disease. A 52-year-old man, stage T1N0,
`progressed to T2N0 in 2 years, and to T2N3 in 4
`years. A 41-year-old woman progressed from
`T1N0 to T3N0 in 13 months. Eight T2N0 or T2N1
`patients progressed to tumor stage within periods
`of 3 months to 12 years (median 62 months). A
`stage T2N0 48-year-old woman progressed to
`T4N3 in 4 years.
`We did not attempt to evaluate the influence of
`various therapies on survival because of multiple
`confounding factors that occur in long-term fol-
`low-up. However, we note that more than 90% of
`T1 patients were treated with topical carmustine.12
`Therapies often used for stage T2 included topical
`carmustine,
`topical corticosteroids,
`low-dose
`methotrexate, and total skin electron beam radia-
`tion (TSEB). For stage T3 these included TSEB,
`local ionizing radiation, and interferon alfa. For
`stage T4 these included low-dose methotrexate,
`interferon alfa, and systemic chemotherapy.
`
`DISCUSSION
`Results of this study demonstrate a strong effect
`of stage, adjusted for age, sex, and race, on sur-
`vival of patients with CTCL. Our cohort of 174 T1
`
`patients had a survival similar to that of an age-,
`sex-, and race-matched population. This is in
`accord with the findings of Kim et al4 who also
`found that the survival of stage IA (T1N0M0)
`patients was similar to that of a race-, age-, and
`sex-matched population. Additionally, we found
`that the survival of T2, T3, and T4 patients was
`significantly shorter than that of a similar control
`population, with T2 showing a moderate, and T3
`and T4 showing a marked decrease in relative
`survival.
`The relative survival of T2 patients with plaque
`stage disease, as defined by histology, was signifi-
`cantly less than that of similar stage patients with
`patch stage disease. This reflects the influence of
`depth of infiltrate and cytologic atypia, and is in
`accord with the findings in a small series reported
`by Marti et al13 in which the survival of T1 and T2
`patients with an infiltrate thicker than 1 mm was
`inferior to that of patients with a thinner infiltrate.
`Our data indicated a marginally significant
`effect of all-inclusive lymph node abnormality on
`survival. The numbers were too small to permit a
`statement as to the effect of different degrees of
`abnormality on survival. Although there was a
`trend for nodal status to be more important in ear-
`lier stage disease this was not statistically signifi-
`cant. Sausville et al3 reported that clinical
`adenopathy was of limited value in assessing prog-
`nosis, but that lymph node histopathologic class
`was a significant determinant of survival. They
`also found that blood and visceral involvement
`was associated with a shorter survival.
`Only 2 (1.1%) of our 174 T1 patients, almost all
`of whom were treated with topical carmustine,
`progressed to a more advanced stage. By contrast,
`11 (9%) of the 122 T1 patients reported by Kim et
`
`IPR2018-00685
`Celgene Ex. 2011, Page 6
`
`

`

`424 Zackheim et al
`
`Journal of the American Academy of Dermatology
`March 1999
`
`al,4 who were treated with topical mechlore-
`thamine or TSEB, progressed to a more advanced
`stage. Nevertheless, there was no significant dif-
`ference in long-term survival for our T1 patients or
`those of Kim et al as compared with age-, sex-, and
`race-matched controls.
`As noted earlier, the percentage of blacks in our
`cohort was not greatly different than that in the
`general population in California. This contrasts
`with other data14 that the relative incidence of MF
`in blacks in the United States, and in San
`Francisco, in the period 1981 to 1984 was approx-
`imately twice that of whites. Multiple factors may
`be involved in the difference between our data and
`that of the earlier period. We also noted previous-
`ly that blacks overall presented with more
`advanced disease than whites. This may reflect a
`delay in seeking medical attention, or a more
`aggressive nature of the disease, and is comparable
`to the situation in prostate cancer.15
`Our data reveal a strong trend to diagnosing
`CTCL at an earlier stage as compared with several
`decades ago. This appears to be in accord with expe-
`rience in other malignancies such as melanoma16
`and prostate cancer.15 The age-adjusted incidence of
`MF in the United States more than doubled from
`1973 to 1984. However, this could represent earlier
`detection rather than a true increase in incidence.
`The analysis of Weinstock and Horm14 suggests
`that the increase is not the result of artifact.
`Nevertheless, it is now some 18 years since those
`data were published and a fresh look appears indi-
`cated. On the other hand, one hopes that earlier
`diagnosis will result in more effective treatment
`and longer survival.
`Our data show a close similarity between the
`relative survival of T3 and T4 patients. This is in
`accord with the findings of Kim et al4 and
`Sausville et al3 in which the unadjusted survival
`curves for T3 and T4 are closely similar. The unad-
`justed survival rate for T3 appears to be better than
`that for T4 in the data of Lamberg et al2 although
`P values are not presented. The unadjusted sur-
`vival for T3 was definitely better than that for T4
`in the data of Toro et al5 but the numbers are small.
`As a rule, in cancer staging systems increasing
`stage correlates with progressively inferior sur-
`vival. However, this does not hold true for the
`majority of studies in CTCL for stages T3 and T4.
`This suggests that the currently used skin T stage
`system is less than ideal, and that a revised T stage
`
`system may be desirable. A proposal for a new T
`stage classification will be the subject of a future
`communication.
`As previously noted, we estimate that 18.8% of
`our patients died from CTCL or related complica-
`tions. This contrasts to some unduly pessimistic
`statements in the literature with regard to the over-
`all prognosis of patients with CTCL. These
`include: “It has not been documented that therapy,
`even when it results in complete clearing of skin
`lesions, changes the ultimately fatal course of the
`disease (MF).”17 “It (MF) is an uncommon, chron-
`ic, itching, usually fatal disease, running a pro-
`gressive course.”18 “The clinical course of MF is
`determined by the stage of disease, and most
`patients eventually die of disease progression or
`infection.”19 We conclude that only 15% to 20% of
`patients with CTCL may be expected to die of
`their disease.
`
`REFERENCES
`1. Koh HK, Charif M, Weinstock MA. Epidemiology and
`clinical manifestations of cutaneous T-cell lymphoma.
`Hematol Oncol Clin North Am 1995;9:943-60.
`2. Lamberg SI, Green SB, Byar DP, Block JB, Clendenning
`WE, Douglass MC, et al. Clinical staging for cutaneous
`T-cell lymphoma. Ann Intern Med 1984;100:187-92.
`3. Sausville EA, Eddy JL, Makuch RW, Fischmann AB,
`Schechter GP, Matthews M, et al. Histopathologic stag-
`ing at initial diagnosis of mycosis fungoides and the
`Sézary syndrome: definition of three distinctive prognos-
`tic groups. Ann Intern Med 1988;109:372-82.
`4. Kim YH, Jensen RA, Watanabe GL, Varghese A, Hoppe
`RT. Clinical stage IA (limited patch and plaque) mycosis
`fungoides. Arch Dermatol 1996;132:1309-13.
`5. Toro JR, Stoll HL, Stomper PC, Oseroff AR. Prognostic
`factors and evaluation of mycosis fungoides and Sézary
`syndrome. J Am Acad Dermatol 1997;37:58-67.
`6. LeBoit PE, McCalmont TH. Cutaneous lymphomas and
`leukemias. In: Elder D, Elenitsas R, Jaworsky C, et al,
`editors. Lever’s Histopathology of the skin. 8th ed.
`Philadelphia: Lippincott-Raven; 1997. p. 805-46.
`7. Bunn PJ, Lamberg SI. Report of the committee on stag-
`ing and classification of cutaneous T-cell lymphoma.
`Cancer Treat Rep 1979;63:725-8.
`8. Wallace AB. The exposure treatment of burns. Lancet
`1951;1:501-4.
`9. Chan CM, Oreglia A. California life expectancy:
`abridged life table for California and Los Angeles
`County, 1989-91. Sacramento: Health and Welfare
`Agency, Department of Health Services, 1993.
`10. Koch SE, Zackheim HS, Williams ML, Fletcher V,
`LeBoit PL. Mycosis fungoides beginning in childhood
`and adolescence. J Am Acad Dermatol 1987;17:563-70.
`11. Weinstock MA, Reynes JF. Validation of cause of death
`certification for outpatient cancers: the contrasting cases
`of melanoma and mycosis fungoides. Am J Epidemiol
`1998;148:1184-6.
`12. Zackheim HS, Epstein EH Jr, Crain WR. Topical car-
`mustine (BCNU) for cutaneous T cell lymphoma: a 15-
`
`IPR2018-00685
`Celgene Ex. 2011, Page 7
`
`

`

`Journal of the American Academy of Dermatology
`Volume 40, Number 3
`
`Zackheim et al 425
`
`year experience in 143 patients. J Am Acad Dermatol
`1990;22:802-10.
`13. Marti RM, Estrach T, Reverter JC, Mascaro JM.
`Prognostic clinicopathologic factors in cutaneous T-cell
`lymphoma. Arch Dermatol 1991;127:1511-6.
`14. Weinstock MA, Horm JW. Mycosis fungoides in the
`United States: increasing incidence and descriptive epi-
`demiology. JAMA 1988;260:42-6.
`15. Haas GP, Sakr WA. Epidemiology of prostate cancer. CA
`Cancer J Clin 1997;47:273-87.
`16. Swerlick MA, Chen S. The melanoma epidemic: Is
`increased surveillance the solution or the problem? Arch
`Dermatol 1996;132:881-4.
`
`17. Lynch PJ. Lymphoma, leukemia, and related diseases.
`In: Sams WM Jr, Lynch PJ, editors. Principles and prac-
`tice of dermatology, 2nd ed. New York: Churchill
`Livingstone; 1996. p. 308.
`18. Arnold HL Jr, Odom RB, James WD, editors. Andrews’
`Diseases of the skin. Philadelphia: WB Saunders; 1990.
`p. 854.
`19. Saleh MN, LeMaistre CF, Kuzel TM, Foss F, Platanias
`LC, Schwartz G, et al. Antitumor activity of DAB389IL-
`2 fusion toxin in mycosis fungoides. J Am Acad
`Dermatol 1998;39:63-73.
`
`ON THE MOVE?
`
`Send us your new address at least six weeks ahead
`
`Don’t miss a single issue of the journal! To ensure prompt service when you change your address,
`please photocopy and complete the form below.
`
`Please send your change of address notification at least six weeks before your move to ensure continued service.
`We regret we cannot guarantee replacement of issues missed due to late notification.
`JOURNAL TITLE:
`Fill in the title of the journal here.
`
`OLD ADDRESS:
`Affix the address label from a recent issue of the journal here.
`
`NEW ADDRESS:
`Clearly print your new address here.
`
`Name
`
`Address
`
`City/State/ZIP
`
`OR FAX TO:
`314-432-1158
`
`OR PHONE:
`1-800-453-4351
`Outside the U.S., call
`314-453-4351
`
`COPY AND MAIL THIS FORM TO:
`Periodical Subscription Services
`Mosby, Inc.
`11830 Westline Industrial Dr.
`St. Louis, MO 63146-3318
`
`IPR2018-00685
`Celgene Ex. 2011, Page 8
`
`